Is artificial intelligence for everyone

What is Artificial Intelligence?

Technology has already overtaken people in some areas. The question now arises: How can companies best use this for themselves? One keyword keeps cropping up in this context: artificial intelligence (AI), also known as artificial intelligence (AI). Machine learning (ML), which is already used by many companies, is part of AI. We provide answers to the most important questions.

Artificial intelligence is the umbrella term for applications in which machines provide human-like intelligence such as learning, judging and problem-solving. The technology of the machine learning (ML) - a branch of artificial intelligence - teaches computers to learn from data and experience and to perform tasks better and better. Sophisticated algorithms can recognize patterns in unstructured data sets such as images, texts or spoken language and use these to make decisions independently.

Definition of AI

Artificial intelligence is the umbrella term for applications in which machines provide human-like intelligence. This includes machine learning or machine learning, natural language processing (NLP) and deep learning. The basic idea is to use machines to approximate important functions of the human brain - learning, judgment and problem solving.

This type of learning enables so-called Natural Language Processing (NLP), among other things. This involves the processing of texts and natural human language, which is used, among other things, by Amazon's Alexa voice service. Deep learning (DL), which uses very deep neural networks with several levels and a large volume of data, is currently seen as the most promising method of machine learning.

In contrast to NLP, the algorithm goes deeper with DL: The machine recognizes structures, can evaluate them and independently improve itself in several forward and backward runs. The algorithm uses several node levels (neurons) in parallel to make well-founded decisions. For example, medicine finds support with deep learning in the early detection of cancer or heart disease and can examine DNA profiles of children for gene markers that indicate type 1 diabetes. In research, deep learning is used, among other things, to evaluate thousands of cell profiles and their active genes or showers of particles that occur when proton beams collide in a particle accelerator. Since this type of learning solves complex, non-linear problems, it is also used in self-driving vehicles, for example, in order to correctly interpret confusing traffic scenes: pedestrians, cyclists, weather, traffic signs or trees - the behavior of road users must be correct, taking into account all possible influencing factors recognized and predicted.

When did artificial intelligence come about?

Since the potential of computers became clear in the 1950s, AI has also developed into a topic that aroused fantasies. In 1970, Marvin Minsky, the "father of AI," declared that machines would be reading Shakespeare in a few years. But nothing like that happened.

Then computers celebrated success after success: First with pure computing power. In 1997 a computer beat the then world chess champion Kasparov in a tournament and in 2011 a computer won the quiz show "Jeopardy". But computers only grew beyond pure computing power when software surprisingly defeated the exceptional Go player Lee Sedol four to one. Because the Asian strategy game Go is much more complex than chess or Jeopardy, in which the computer basically only has to understand questions and search for answers in a database. The computer won the Go games because it is much more than a fast computer - the software learns. A stronger version later even learned the game from scratch and won against its predecessors.

How do machines learn?

Machine learning takes place either through training using a data set with already known outputs (monitored), or algorithms have to recognize patterns in data themselves (unsupervised). Learning through reward and punishment (reinforced) is also possible, in which the algorithm automatically detects whether the learning component uses the entire system (reward) or not (punishment). The data is either available in a structured form, for example in tabular form, or unstructured as text, images or language - as in emails or social media posts. Machine learning can process any data, which is a huge advantage.

In which areas can AI be used?

AI is interesting for all industries that generate large amounts of data. For example, for manufacturing companies where suppliers, sensors in the machines and the ERP system can provide a lot of data. Self-learning algorithms support quality control and provide forecasts for predictive maintenance of the machines. In this way, companies avoid production downtimes and minimize storage costs, to name just a few examples.

In the healthcare sector, too, there are almost unlimited possibilities for the use of AI thanks to medical image analysis or robot-assisted surgery. In every industry, ideas are currently emerging that often lead to significant efficiency gains, since repeatable tasks in processes run automatically. This gives people more time for strategically important and creative tasks. But AI also leads to new business models - for example, when a company no longer sells machines, but instead sells their performance.

What are the benefits for companies with AI?

Artificial intelligence simplifies work processes, enables more precise forecasts and creates new data-based business models. It allows faster decisions on a better database, increases the ability of companies to adapt to market changes through real-time information and predictions beyond human capabilities. So for companies, AI creates far more than efficiency - it is a key to increased competitiveness.

How can AI be used in companies?

According to IDC, 94 percent of companies believe that machine learning is the key to significant competitive advantage. Not without reason - because artificial intelligence strengthens companies in terms of productivity, flexibility and creates new business value, for example through intelligent chatbots in service.

Through personalized customer service, algorithms learn from direct interaction with customers and respond more precisely to their needs. Here, too, SAP relies on intelligent processes and supports its customers with appropriate services and applications. This includes:

  • SAP CoPilot, a digital chat assistant that uses questions and answers to help users achieve their goals faster.
  • SAP Service Ticket Intelligence. The application automatically categorizes customer reports, prioritizes the tasks to be processed for employees or provides suggestions for answers to standard inquiries.
  • SAP Customer Retention predicts customer behavior.
  • SAP Resume Matching helps to find the most suitable job applicants from a large number of documents.
  • SAP Brand Impact helps companies measure their own brand influence and the success of investments in sponsorship and advertising. The solution records, for example, how often your own company logo appears in video material, where and in what size.

How does machine learning expand human capabilities?

A self-learning algorithm recognizes even the smallest changes and can assess the effects, especially where people could make mistakes, such as in quality control in production. Inspections of long pipelines, for example, can be carried out in an impressive manner using drones and satellites. This is also of great importance when it comes to data security. Machine learning quickly locates anomalies in transactions and processes, detects attempts at bribery and effectively protects against hacking. Everyday processes are also simplified. If a train unexpectedly fails, an algorithm adjusts the individual travel planning with a context-sensitive application: the customer sees alternative routes online or in an app that will get them to their destination as quickly as possible despite the failure.

How can my company integrate machine learning?

Every company should know which machine learning options are already available in the solution used. Because your own machine learning expertise is not necessary for every intelligent application. Machine learning is already integrated in SAP products such as Concur or SAP S / 4HANA and with the help of the SAP Leonardo Machine Learning Foundation, every SAP partner or SAP user can link and implement ready-made services, but also their own models for creating intelligent ones Use applications.

The basis for the development of intelligent applications of all kinds, their provision and high-performance operation is provided by the SAP cloud platform. The technology for machine learning is therefore available and many companies have already found approaches for its meaningful use.

What does the future look like - what will AI and machine learning bring?

Machine learning can process structured and unstructured data well. It is conceivable and already partially feasible that machines interact without instructions, carry out highly context-sensitive actions, draw their own conclusions and adapt their behavior. They focus the employees' valuable time on the essentials: creativity and innovation. When it comes to creating your own ideas, AI is currently reaching its limits. In addition, digital assistants and bots will support us in our day-to-day work in the future. New organizational roles will also be created to show ways of successful collaborative work between man and machine.

Additional Information:

You have further questions? Then take a look at our other articles on machine learning.

Image via Shutterstock